Thursday, June 4, 2009

Ars Technica: 'RIAA v. Jammie Thomas, round two: an in-depth preview'

Nate Anderson of Ars Technica has written a characteristically solid preview of the June 15 re-trial of accused peer-to-peer infringer Jammie Thomas. His summary of the record labels' case:

The RIAA's case is less convoluted [than Thomas']. On February 21, 2005, MediaSentry detected 1,702 songs being shared by a KaZaA user with the IP address of The username on the account was "tereastarr@KaZaA." MediaSentry downloaded some of these songs, logged the whole event, and then sent instant messages to the account in question, warning them about copyright infringement. (These messages constitute one of the RIAA's exhibits.)

With the info in hand, the recording industry then filed a "Doe" lawsuit and obtained judicial permission to file a subpoena on Charter Communications. Charter said that the subscriber at the IP address in question, at the time in question, was one Jammie Thomas.

Fair enough, but we've seen plenty of errors creep into this process. Does the recording industry actually have any corroborating evidence?

It does. A later forensic search of Thomas' machine turned up cached browser pages showing that Thomas had used "tereastarr" as her username on EA Sports,, and Yahoo—and the screenshots to prove it are also on the RIAA's exhibit list for the case.

Just a couple quibbles. Anderson writes:
In addition to all the other claims, Thomas wants the MediaSentry evidence thrown out altogether, a strategy that looks more likely to succeed based on Minnesota law governing private investigators. (MediaSentry obtained no such PI licenses in the states it operated.)
I disagree that Thomas' attempt to exclude the MediaSentry evidence has much chance of success. First, I have grave doubts whether the Minnesota private investigator statute even applies to activity such as that performed by MediaSentry: simply joining a p2p network and recording what it saw there. Second, it's quite a stretch to say that MediaSentry "operated" in Minnesota. MediaSentry was headquartered in Maryland, and I believe it performed its work from there -- no need to set foot in Minnesota to note that an IP address in Brainerd (Thomas' home) was offering songs on Kazaa. (Do I "operate" in Mongolia if I surf to a web site whose server is locate there?) Third, there is no exclusionary rule in federal civil cases. So, even if MediaSentry had failed to comply with Minnesota PI regulations, that would not render inadmissible the evidence it collected. More here and here.

My other quibble is over this passage:
It sounds a bit like the RIAA can't let go of its "making available==actual distribution" theory, even after a judge tossed out the first verdict against Thomas on precisely this issue.

This attitude is confirmed later in the filing, where the RIAA says that "an inference that a distribution actually took place may be made where a defendant has completed all necessary steps for the distribution of copyrighted sound recordings to other users on a peer-to-peer network, without license from the copyright owners." That's just a fancy way of describing the making available theory, and the fact that it's still being trotted out in the Eighth Circuit, in a retrial, is... interesting.

I can't deny that this whole thing is "interesting." But the labels' persistence in pressing the "making available" argument is hardly surprising, and entirely legitimate. They believe they are correct (as do I), and are making this argument (despite the fact that the court has previously rejected it) to preserve it for appeal.

My quibbles aside, a very solid piece. Read the whole thing.


  1. Previous comments seems to have bounced. Ugh. Anyway, I beg to differ with both you and Nate on making available. It looks to be like RIAA is actually backing off its previous stance on making available, which was the argument that making available was itself a violation fo the distribution right. (Not merely tantamount to distribution, but a form of distribution.) Here (in the passage Nate quotes), they seem to be arguing instead that one can draw an inference of ("actual") distribution from the making available of a file. This seems significant, and quite a tenable position. I have little doubt that RIAA has not given up its previous position that MA=distribution, but they seem to recongize they need an alternative int he Thomas case, where that argument has already been rejected.

  2. @ Anonymous 7:38:

    The labels are not backing off the "making available" argument (thought they know the court has already rejected it). They have proposed *both* a making available instruction (Proposed #20), *and* an "inference" instruction like the one you describe (#19). I don't know whether the court will give #19 (though I think it's unlikely).


Comments here are moderated. I appreciate substantive comments, whether or not they agree with what I've written. Stay on topic, and be civil. Comments that contain name-calling, personal attacks, or the like will be rejected. If you want to rant about how evil the RIAA and MPAA are, and how entertainment companies' employees and attorneys are bad people, there are plenty of other places for you to go.